Thermal Loss and Soldering Effect Study of High Q Antennas

Aalborg Universitet

Thermal loss and soldering effect study of high-Q antennas in handheld devices

Bahramzy, Pevand; Jagielski, Ole; Pedersen, Gert Frølund

Published in:Antennas and Propagation (EuCAP), 2013 7th European Conference on

Publication date:2013

Document VersionPublisher's PDF, also known as Version of record

Link to publication from Aalborg University

Citation for published version (APA):Bahramzy, P., Jagielski, O., & Pedersen, G. F. (2013). Thermal loss and soldering effect study of high-Qantennas in handheld devices. In Antennas and Propagation (EuCAP), 2013 7th European Conference on (pp.878 - 881). IEEE. http://ieeexplore.ieee.org/xpl/login.jsp?tp=&arnumber=6546408

General rightsCopyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright ownersand it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights.

? Users may download and print one copy of any publication from the public portal for the purpose of private study or research. ? You may not further distribute the material or use it for any profit-making activity or commercial gain ? You may freely distribute the URL identifying the publication in the public portal ?

Take down policyIf you believe that this document breaches copyright please contact us at [email protected] providing details, and we will remove access tothe work immediately and investigate your claim.

Downloaded from vbn.aau.dk on: December 30, 2021

https://vbn.aau.dk/en/publications/9ea51da7-dc23-4dfe-9896-3b3803656d22

http://ieeexplore.ieee.org/xpl/login.jsp?tp=&arnumber=6546408

Thermal Loss and Soldering Effect Study ofHigh-Q Antennas in Handheld Devices

Pevand Bahramzy∗, Ole Jagielski∗, Gert F. Pedersen†,

∗Molex Interconnect - Antenna Business Unit (ABU), 35 Lindholm Brygge, Noerresundby, 9400, Denmark;Email: pevand.bahramzy, [email protected]

†Section of Antennas, Propagation and Radio Networking (APNet), Department of Electronics Systems,Faculty of Engineering and Science, Aalborg University, DK-9220, Aalborg, Denmark

Email: [email protected]

Abstract—High-Q antennas are attractive because, besides be-ing narrow-band, they have the advantage of being more compactand therefore occupy less volume in a mobile device. However,they can become very lossy especially at lower frequencies. Inthis paper it is investigated how low a thermal loss, in a verygood conductor as copper, may be achieved. The effect fromsolderings on the antenna efficiency is also investigated and theeffect has shown to be small. The resistance value, based on theextra loss due to the solderings, is estimated to be 0.25 Ohm.It is also shown that two different high-Q antennas, having thesame Q value, can have difference in efficiency. Furthermore, itis discussed why it is so difficult to compare the electrical sizeand volume of different antenna types.

Index Terms—High-Q antenna; soldering; thermal loss; effi-ciency

I. INTRODUCTION

According to Long Term Evolution (LTE) standard, themobile terminals will have to operate on more than 20 bandsover frequencies between 700 MHz and 2700 MHz. It iswell known that the antenna design is in general challengingdue to the fundamental limitation of antennas [1], [2]. Withthe introduction of LTE, the challenge gets even bigger foran antenna designer to cover the whole frequency spectrumdue to size constraints and very limited available space forantennas in close proximity to components such as camera,speaker, batterie and other hardware. One way to cover thewhole spectrum is to use reconfigurable antennas. Thesereconfigurable antennas can be designed to be very narrow-band, since they only need to cover one channel instead ofa full band, and tuned to resonate at different frequencies.Channels in LTE are between 1.4 MHz and 20 MHz [3]. Thisway, the reconfigurable antenna can cover a large bandwidth.A tunable antenna has the advantage that it can reuse its entirevolume at different operating bands so the physical size of theantenna can be reduced [4]. This narrow-band advantage oftunable antennas allows for designs with a high Quality factor(Q).

Pevand is now working for Intel Mobile Communications

High-Q antennas have relative high current and field densityper area. Due to the high current density associated with high-Q many challenges, such as more loss in the antenna conduc-tor, interconnection, carrier and tuning/matching components,are introduced.

Fig. 1. Geometry of the antenna structure. (Right) antenna element on topof the PWB. (Left) the feeding point connected to a coaxial cable.

One of the main concerns with an high-Q antenna is theundesired high thermal loss in the conductor. Solder can,due to worse conductivity than copper, cause more thermalloss in the conductor. In this paper it is investigated: 1) howgood an efficiency can be achieved with an high-Q antennamade of copper, and 2) how much the antenna thermal lossincreases due to the solderings at high current paths of anhigh-Q antenna. The antenna designs for the soldering effectand thermal loss investigations are presented in Section II.The results are shown and discussed in Section III and finallyconclusion is disclosed in section IV.

978-88-907018-3-2/13 ©2013 IEEE

2013 7th European Conference on Antennas and Propagation (EuCAP)

878

II. ANTENNAS FOR THE SOLDERING EFFECT ANDTHERMAL LOSS STUDY

An Inverted F Antenna (IFA), which is widely used inmobile phone industry because of its integrated and low profiledesign, is designed for the soldering effect and thermal lossstudy. A patch antenna, having same Q as the IFA, is designedfor the thermal loss comparison.

A. Inverted F Antenna

Figure 1 shows a sort of meandered IFA on top of a PWB.The Printed Wire Board (PWB) has the total dimensions of100x40 mm2 and the IFA has the dimensions 40x20x5 mm3.The PWB and the antenna is designed as one structure, andthe antenna part is then bent in order to create the IFA. In thisway, soldering is only needed at the feed point (se Figure 1).The soldering at the feed should not cause much loss as theimpedance is relatively high here - some 50Ω.

Fig. 2. Log Magnitude impedance plot of the IFA antenna.

The antenna is selfmatched, so no matching componentsare used. This in order to see how good an efficiency canbe achieved with an antenna consisting of just copper. Airis preferred between the antenna element and the PWB inthis study, because the effect of carrier loss is undesirable.However, a support material is necessary in order to make theantenna stable. Rohacell 31 HF [5], εr=1.050 and tan δ <0.0002, is used as support material because of very low loss.The investigation is made at frequencies in the 700 MHz bandbecause this band is the toughest band in terms of loss. Theantenna impedance is seen in Figure 2, where the abs(S11)=-6dB matched bandwidth is 14 MHz and the Q = 60.

As seen in Figure 3, different positions at the short of theantenna are soldered due to the high current density there.The antenna efficiency is measured with and without thesolderings in order to see how much the loss increases dueto the solderings. All the measurements are done using an RFchoke (se Figure 4) in order to avoid efficiency contributionfrom currents flowing on measurement cable.

B. Patch Antenna

A patch antenna is designed in order to find out if a betterefficiency can be achieved compared to the IFA when the two

Fig. 3. Different soldering positions at the short of the IFA.

Fig. 4. Measurement setup of the mock-up with IFA antenna in the anechoicchamber.

antenna Q are comparable. Figure 5 shows an quadratic patchantenna. This patch antenna has the dimensions 70x70x10mm3. The whole patch is designed as one structure and thenbent, so the solder is only used at the feed point (se Figure5).

The patch antenna is also selfmatched (no matching com-ponents) in order to see how good an efficiency can beachieved with the patch antenna build of copper. Small brick of

Fig. 5. Geometry of the patch antenna structure.


879

Rohacell 31 HF is used, as supoort material between the twopatch plates, in order to make the patch antenna stable. Thepatch antenna is also designed to resonate at 700 MHz band(resonance is about 20 MHz lower in frequency compared tothe IFA resonance). The patch antenna is designed to have thesame impedance bandwidth (14 MHz) and Q (60) as the IFA(see Figure 6).

Fig. 6. Log Magnitude impedance plot of the patch antenna.

III. MEASUREMENT RESULTS

TABLE IMEASURED EFFICIENCY RESULTS FOR DIFFERENT SOLDERING POSITIONS

AT SHORT OF THE IFA.

Frequency in MHz735 740 749.5

Solder only at feed

S11 (dB) -6.4 -12.2 -6.2Mismatch loss (dB) 1.1 0.3 1.2Thermal loss (dB) 0.2 0.6 1.4

Total Efficiency (dB) -1.4 -0.9 -2.6

Solder at short corner



Solder at both short corners



Short cut and soldered at corners



Short cut and soldered all the way



TABLE I shows the measured efficiency results for differentsoldering positions at short of the IFA. The thermal lossis affected by how good the antenna is matched. There isa tendency of thermal loss increasing when mismatch lossdecreases and vice versa (se TABLE I). This inverse relationis not one to one. Therefore, for each case S11, mismatchloss, thermal loss and total efficiency measurements are shown.The first measurement shows how good an efficiency whichcan be achieved with an antenna made of just copper. Themeasurement shows thermal loss of 0.6 dB with a mismatchloss of 0.3 dB. If the antenna was fully matched, then thethermal loss would probably be slightly higher. However, itseems that an antenna efficiency of -0.9 dB can be achievedwith mock-up made of only copper.

The next four measurements show the effect of solderingsat the short of the antenna. The solderings have in generallittle effect on the thermal loss. An increase of up to 0.3 dBis seen in the thermal loss due to the solderings. A resistancevalue, based on the loss due to the solderings, can be estimatedin order to see how much ohmic loss it corresponds to. Thisresistance value can then be used in simulations to equivalatethe loss due to the solderings. The resistance value, using ADSsimulations, is estimated to be 0.25 Ohm.

TABLE IIMEASURED EFFICIENCY RESULTS OF THE PATCH ANTENNA.

Frequency in MHz711 718 724

Quadratic patch



TABLE II shows the measured efficiency results of the patchantenna. The results in TABLE II are compared to the firstmeasurement in TABLE I. The patch antenna shows to havebetter efficiency than the IFA antenna on top of a PWB, eventhough they have equal Q values. The measurements wererepeated in order to verify the correctness of the results, butwith the same outcome. The patch antenna has thermal loss of0.4 dB with no mismatch loss, where the IFA has thermal lossof 0.6 dB and mismatch loss of 0.3 dB. If the IFA antennawas fully matched, then the thermal loss would probably beslightly higher, due to the tendency of increasing thermal losswith decreasing mismatch loss.

IV. CONCLUSION

One of the objectives, in this paper, was to study the effect ofsolderings on the antenna efficiency. The measurements haveshown a degradation in thermal loss, up to 0.3 dB, due to thesolderings. It seems that solderings at high current paths of ahigh-Q antenna have very little effect on the thermal loss. Theresistance value, based on the extra loss due to the solderings,is estimated to be 0.25 Ohm.

Another objective was to show how good an efficiency thatcan be achieved with an high-Q antenna made of a very goodconductor as copper. It seems that no better than some -0.9 dBof efficiency can be achieved with an IFA antenna on top of aPWB. However, the patch antenna, having the same Q value,shows to have efficiency of around -0.4 dB. The difference inefficiency is significant between the two antenna types.

One way to explain this difference in efficiency, is to lookat the structure of the two antenna types. The IFA antennawill have high current density per area due to its meanderingstructure. The patch antenna on the other hand has a simplestructure and is therfore expected to have lower current densityper area, which imply lower loss. The mock-up with the IFAantenna, due to the meandering structure, has longer currentpath compared to the patch antenna. Long current path alsocauses more loss in the structure.


880

The patch antenna and IFA on top of a PWB are comparedin terms of Q value only, which can be an unfair comparison.The comparison will be more equitable if electrical size,based on antenna volume of both antenna types, are alsotaken into account, because Q and antenna electrical size haveinverse relationship (Q grows rapidly as antenna electrical sizedecreases). The problem is how to compare the electrical sizeand volume of different antenna types. In the following it willbe discussed why it is so immensely hard to compare the twodifferent antenna types in terms of electrical size and volume.

The IFA is a monopole antenna on top of a PWB, where thePWB is the main radiator and the IFA acts more as a coupler,specially at low band frequencies (700 MHz). The volumeof an antenna, in a mobile phone, is typically defined as thearea between the antenna element and the PWB. However, thisdefinition is vague since it is very difficult to define what isthe antenna and what is not. The volume of the patch antennacan be explicitly expressed, but cannot be directly comparedto the IFA antenna volume, because the patch antenna volumeis the whole structure volume (70x70x10mm3), where the IFAantenna volume is the area between the antenna element andthe PWB. The patch antenna has a dipole mode, meaning thatthe currents run in phase on both patch plates. On the contrary,the IFA and the PWB has opposite phase currents.

While the electrical length of a mobile phone is typicallycalculated taking the length + width of the PWB into account,it becomes more blurry to express the combined electricallength when adding the antenna into the equation, becauseincreasing e.g. the length of the PWB will mean increasedelectrical size. However, this does not imply lower Q, since thePWB moves away from its resonance frequency and thereforethe combined PWB + IFA will result in lower bandwidth.

REFERENCES

[1] H. Wheeler, Fundamental limitations of small antennas,Proceedings of the IRE, vol. 35, no. 12, pp. 1479 1484,dec. 1947.

[2] L. J. Chu, Physical limitations of omni-directional an-tennas, Journal of Applied Physics, vol. 19, no. 12, pp.11631175, 1948.

[3] Evolved Universal Terrestrial Radio Access (E-UTRA);User Equipment (UE) radio transmission and reception,3GPP Std. TS 36.101, http://www.3gpp.org/ftp/Specs/html-info/36101.htm.

[4] A.C.K. Mak, C.R. Rowell, R.D. Murch and C.-L. Mak,Reconfigurable Multiband Antenna Designs for Wire-less Communication Devices, IEEE Transaction AntennasPropag., vol 55, No 7, pp. 1919-1928, July 2007.

[5] Rohacell; available at: http://www.rohacell.com/product/rohacell/en/products-services/rohacell-hf/pages/default.aspx


881

Documents

Thermal Loss and Soldering Effect Study of High Q Antennas